battery-analysis

# Battery Analysis ## Critical Rules 1. **Always specify the battery chemistry** — Li-ion, Na-ion, solid-state, etc. Voltage windows, stability criteria, and analysis methods differ significantly. 2. **Cycling data must be parsed from the native potentiostat format** — CSV exports lose metadata (mass loading, cell area, protocol details). Use `galvani` for Biologic .mpr, `cellpy` for Arbin/Maccor. 3. **Report capacities normalized correctly** — mAh/g (gravimetric) requires accurate active mass. mAh/cm² (areal) requires electrode area. Always state which. 4. **Voltage profiles shift with C-rate** — don't compare 0.1C and 1C profiles directly without noting the rate. 5. **Coulombic efficiency is the most important cycling metric** — CE > 99.9% is typically needed for practical cells. Plot CE on a separate y-axis with expanded scale. 6. **For computational screening**, always check electrochemical stability window (voltage vs Li/Li+) using the Materials Project phase diagram, not just thermodynamic stability. ## Potentiostat Data Parsing ### Biologic (.mpr files) ```python from galvani import BioLogic mpr = BioLogic.MPRfile("cycling_data.mpr") df = mpr.data # Common columns: time/s, Ewe/V, I/mA, capacity/mA.h, cycle number print(df.columns.tolist()) # Basic cycling plot import matplotlib.pyplot as plt fig, ax1 = plt.subplots(figsize=(10, 5)) ax1.plot(df["capacity/mA.h"], df["Ewe/V"]) ax1.set_xlabel("Capacity (mAh)") ax1.set_ylabel("Voltage (V)") plt.savefig("voltage_profile.png", dpi=150) ``` ### Arbin / Maccor (via cellpy) ```python from cellpy import cellreader # Load raw data — cellpy handles Arbin .res and Maccor formats cell = cellreader.CellpyCell() cell.from_raw("arbin_data.res") # Get summary per cycle summary = cell.cell.summary print(f"Cycles: {len(summary)}") print(f"First discharge capacity: {summary.iloc[0]['discharge_capacity']:.1f} mAh/g") # Get step data for voltage profiles steps = cell.cell.steps # Capacity vs cycle number fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5)) ax1.plot(summary.index, summary["discharge_capacity"], 'o-') ax1.set_xlabel("Cycle") ax1.set_ylabel("Discharge Capacity (mAh/g)") ax2.plot(summary.index, summary["coulombic_efficiency"] * 100, 'o-') ax2.set_xlabel("Cycle") ax2.set_ylabel("Coulombic Efficiency (%)") ax2.set_ylim([99, 101]) plt.tight_layout() plt.savefig("cycling_summary.png", dpi=150) ``` ## Electrochemical Impedance Spectroscopy (EIS) ```python from impedance import preprocessing from impedance.models.circuits import CustomCircuit # Load EIS data (frequency, Z_real, Z_imag) f, Z = preprocessing.readCSV("eis_data.csv") # Ignore points below 0 (instrument artifacts) f, Z = preprocessing.ignoreBelowX(f, Z) # Define equivalent circuit # R0 = ohmic resistance, R1-CPE1 = charge transfer, W1 = Warburg diffusion circuit = "R0-p(R1,CPE1)-W1" initial_guess = [10, 100, 1e-4, 0.8, 500] circuit_model = CustomCircuit(circuit, initial_guess=initial_guess) circuit_model.fit(f, Z) print(f"Ohmic resistance: {circuit_model.parameters_[0]:.2f} Ω") print(f"Charge transfer resistance: {circuit_model.parameters_[1]:.2f} Ω") # Nyquist plot fig, ax = plt.subplots(figsize=(8, 8)) ax.plot(Z.real, -Z.imag, 'o', label="Data") Z_fit = circuit_model.predict(f) ax.plot(Z_fit.real, -Z_fit.imag, '-', label="Fit") ax.set_xlabel("Z' (Ω)") ax.set_ylabel("-Z'' (Ω)") ax.set_aspect("equal") ax.legend() plt.savefig("nyquist.png", dpi=150) ``` ## Computational Battery Screening ### Electrode voltage and capacity prediction ```python from mp_api.client import MPRester from pymatgen.apps.battery.insertion_battery import InsertionElectrode from pymatgen.analysis.phase_diagram import PhaseDiagram from pymatgen.entries.compatibility import MaterialsProjectCompatibility with MPRester() as mpr: # Get all entries in the Li-Mn-O system entries = mpr.get_entries_in_chemsys(["Li", "Mn", "O"]) # Apply compatibility corrections compat = MaterialsProjectCompatibility() entries = compat.process_entries(entries) # Build insertion electrode (Li into MnO2 framework) # This calculates voltage profile and theoretical capacity ie = InsertionElectrode.from_entries(entries, working_ion="Li") print(f"Average voltage: {ie.get_average_voltage():.2f} V vs Li/Li+") print(f"Theoretical capacity: {ie.get_capacity_grav():.0f} mAh/g") print(f"Energy density: {ie.get_energy_grav():.0f} Wh/kg") ``` ### Electrochemical stability window ```python from mp_api.client import MPRester from pymatgen.analysis.phase_diagram import PhaseDiagram with MPRester() as mpr: # Check stability of a solid electrolyte against Li metal entries = mpr.get_entries_in_chemsys(["Li", "P", "S"]) pd = PhaseDiagram(entries) # Get stability window — voltage range where electrolyte doesn't decompose # This is critical for solid-state battery electrolyte selection from pymatgen.analysis.phase_diagram import GrandPotentialPhaseDiagram # Sweep Li chemical potential to find stability window for voltage in [0, 1, 2, 3, 4, 5]: mu_li = -voltage # μ_Li = -eV vs Li/Li+ # Check if the material is stable at this potential # ... (use GrandPotentialPhaseDiagram) ``` ## Physics-Based Modeling (PyBaMM) ```python import pybamm # Single Particle Model — fast, good for initial analysis model = pybamm.lithium_ion.SPM() # Or full Doyle-Fuller-Newman model — more accurate # model = pybamm.lithium_ion.DFN() # Use built-in parameter set param = pybamm.ParameterValues("Chen2020") # Simulate 1C discharge sim = pybamm.Simulation(model, parameter_values=param) sim.solve([0, 3600]) # 1 hour = 1C # Plot sim.plot(["Terminal voltage [V]", "Current [A]", "Negative particle concentration [mol.m-3]", "Positive particle concentration [mol.m-3]"]) ``` ## Common Battery Metrics | Metric | Good Target | How to Calculate | |---|---|---| | Gravimetric capacity | >150 mAh/g (cathode) | Discharge capacity / active mass | | Coulombic efficiency | >99.9% | Discharge capacity / charge capacity × 100 | | Rate capability | >80% at 2C vs 0.1C | Capacity at high rate / capacity at low rate | | Cycle retention | >80% after 500 cycles | Capacity at cycle N / capacity at cycle 1 | | ICE (1st cycle) | >85% | 1st discharge / 1st charge × 100 | | Voltage hysteresis | <0.2 V | Average charge V - average discharge V | | ASR (impedance) | <20 Ω·cm² | From EIS fitting | ## Common Pitfalls - **Mass loading errors** propagate directly to capacity — verify active mass carefully. - **Formation cycles** (first 1-3 cycles) should be analyzed separately — SEI formation causes irreversible capacity loss. - **Temperature matters** — always report cell temperature. Capacity increases ~1%/°C. - **Calendar aging** — cells degrade even when not cycling. Note rest periods. - **Two-electrode vs three-electrode** — coin cell data mixes cathode and anode contributions. Use reference electrode for single-electrode analysis. ## Reference See `references/cellpy-guide.md` for detailed potentiostat data handling. See `references/battery-screening.md` for computational screening workflows.